Causal Structure and Representation Learning with Biomedical Applications

TL;DR

This paper explores integrating causal inference with representation learning to better understand complex biomedical data, focusing on causal discovery, multi-modal data integration, and optimal perturbation design.

Contribution

It proposes a new framework combining causal structure learning with representation learning tailored for multi-modal biomedical data analysis.

Findings

Framework for causal discovery using observational and perturbational data

Method for learning causal variables from multi-modal data

Approach for designing optimal perturbations in biomedical systems

Abstract

Massive data collection holds the promise of a better understanding of complex phenomena and, ultimately, better decisions. Representation learning has become a key driver of deep learning applications, as it allows learning latent spaces that capture important properties of the data without requiring any supervised annotations. Although representation learning has been hugely successful in predictive tasks, it can fail miserably in causal tasks including predicting the effect of a perturbation/intervention. This calls for a marriage between representation learning and causal inference. An exciting opportunity in this regard stems from the growing availability of multi-modal data (observational and perturbational, imaging-based and sequencing-based, at the single-cell level, tissue-level, and organism-level). We outline a statistical and computational framework for causal structure and representation learning motivated by fundamental biomedical questions: how to effectively use observational and perturbational data to perform causal discovery on observed causal variables; how to use multi-modal views of the system to learn causal variables; and how to design optimal perturbations.